Lantibiotics are gene-encoded, ribosomally synthesized derived peptides that have attracted widespread scientific attention in recent years, not only as promising safe and natural food additives but also as potential chemotherapeutic agents. Lantibiotics are produced by a large number of gram-positive bacteria and are as such considered members of a group of bacterial toxins called bacteriocins. The original, and most intensively studied, lantibiotic is the nisin lantibiotic.
Nisin is a polycylic, 34 amino acid peptide with antibacterial activity against a range of gram-positive bacteria and a small number of gram-negative bacteria. These include food-borne pathogens such as staphylococci, bacilli, clostridia and mycobacteria. Nisin is FDA approved with a long record of safe use (Delves Broughton, 1990) and is one of only a few bacteriocins to have been applied commercially (Twomey et al., 2002). Nisin's commercial use in the food industry stems from its ability to suppress gram-positive spoilage and other pathogenic bacteria. It also possesses low anti-gram negative activity, which increases when combined with other hurdles e.g. high pressure.
Nisin has also been proven to inhibit the pathogenic bacteria responsible for bovine mastitis including Streptococcus agalactiae, Strep. dysgalactiae, Strep. uberis and Staphylococcus aureus. Bovine mastitis is an inflammation of the udder that is both persistent and costly to treat. Consequently, in recent years nisin has been incorporated into a number of commercial products that are used as an alternative treatment for bovine mastitis (Sears et al., 1992; Wu et al., 2007). For example, Immucell produce Wipe Out®, used to clean and sanitize the teat area before and after milking. This successfully reduced levels of the mastitis pathogens Staph. aureus (99.9%), Strep. agalactiae (99.9%), E. coli (99%), Step. uberis (99%) and Klebsiella pneumoniae (99%) in experimental exposure studies (J. Dairy Sci 75:3185-3192). Mast Out®, a nisin-based treatment for mastitis in lactating cows has been shown to give statistically significant cure rates in an experimental field trial involving 139 cows with subclinical mastitis. Similarly, another lantibiotic, lacticin 3147 , has been successfully incorporated into a teat seal product with a view to protecting the seal during the ‘drying-off’ period—nisin derivatives could be similarly employed.
Although not currently being used to treat any particular diseases, there is a great deal of potential for nisin use based on the fact that it inhibits a number of pathogenic microbes. The effectiveness of nisin against enterococci and staphylococci and mycobacteria has been shown, as has its activity against Clostridium difficile. 
Lantibiotics are generally regarded as possessing poor anti-gram-negative activity. This insensitivity is thought to due to an inability to passage across the outer membrane of the gram-negative cell wall, thus limiting access to lipid II. However, this general trend is not always strictly true. In fact, it has been established that in its purified form, nisin Z exhibits activity against other gram-negative microbes such as Escherichia coli and S. aureus. Both nisin A and Z exhibited activity against two antibiotic resistant strains of Neisseria gonorrhoeae and Helicobacter pylori. Another gram-negative bacteria that has shown susceptibility to nisin, whether alone or in combination with other antimicrobials, is Pseudomonas aeruginosa. In yet another application nisin has also been shown to have potential as a contraceptive.
It has been demonstrated in laboratory settings that bacteria can become resistant to nisin, e.g. serial exposure of a penicillin-susceptible strain of Strep. pneumoniae to nisin (1 mg/L) in liquid culture resulted in the rapid appearance of stable nisin-resistant mutants in which the minimum inhibitory concentration (MIC) increased from 0.4 to 6.4 mg/L (Severina, 1998). In these spontaneous mutants, resistance correlates with cell envelope changes such as alterations in membrane charge and fluidity (Li, 2002; Verheul, 1997), cell wall thickness (Maisnier-Patin, 1996), cell wall charge (Mantovani, 2001; Abachin, 2002; Bierbaum, 1987) and combinations thereof (Crandall, 1998), arising following direct exposure to a low level of lantibiotic or as part of an adaptive response to another stress (van Schaik, 1999). The specific mechanism(s) by which cells become resistant to nisin is not well understood although it is apparent that variations in the lipid II content are not responsible (Kramer, 2004). Genetic loci associated with the development of enhanced nisin resistance (Cotter, 2002; Gravesen, 2004; Gravesen, 2001), or an innate tolerance of nisin, have been identified (Peschel, 1999; Abachin, 2002; Cao, 2004). In the latter example the cell envelope charge would seem to be the most important consideration. While this has not as yet impacted on the application of nisin in the food industry, it has implications for applications in the future and the potential of nisin as a clinical antimicrobial. This however does point to the importance of identifying further antimicrobials including variants of existing antimicrobials, to overcome resistance problems.
Studies investigating the mode of action of lanitbiotics have revealed the membrane-bound peptidoglycan precursor lipid II to be the docking molecule for the nisin lantibiotic. The binding of nisin to lipid II facilitates two bactericidal activities, namely, membrane pore formation and the inhibition of peptidoglycan biosynthesis (Bonelli et al., 2006; Breukink et al., 1999; Brotz et al., 1998; Wiedemann et al., 2001). The dual activity of nisin is thought to be due to the presence of two-structural domains, located at the N- and C-termini respectively. The N-terminal domain contains three post-translationally incorporated (β-methyl) lanthionine rings (rings A, B, and C) and is linked to the C-terminal rings (rings D and E) by a flexible region, or hinge. This hinge consists of three amino acids (Asn20-Met21-Lys22; as illustrated in FIG. 1). It has been established that the A, B and C rings of the N-terminal form a ‘cage’, that facilitates binding to the pyrophosphate moiety of lipid II, thus interfering with the process of cell wall synthesis (Hsu et al., 2004). This binding in turn enhances the ability of the C-terminal segment, containing rings D and E, to form pores in the cell membrane, resulting in the rapid efflux of ions and cytoplasmic solutes, such as amino acids and nucleotides, into the cell (Wiedemann et al., 2001).
To date, six natural variants of nisin have been identified. These variants are nisin A (Kaletta and Entian, 1989), nisin Z (Mulders et al., 1991), nisin Q (Zendo et al., 2003) and nisin F (de Kwaadsteniet et al., 2007) which are produced by Lactococcus lactis species, while nisin U and nisin U2 are produced by Streptococcus uberis (FIG. 1) (Wirawan et al., 2006). The diversity of these natural variants highlights the ability of certain residues and domains within the molecule to tolerate change. However, comparisons between closely (e.g. subtilin) and more distantly related (e.g. epidermin) lantibiotics revealed that highly conserved elements, with essential structure/function roles, also exist. Despite the relatively plastic nature of the nisin peptides, of the bioengineered derivatives of nisin that have been generated and characterized to date, only two (T2S and M17Q/G18T) display increased activity against at least one gram-positive bacteria, and even then, activity is enhanced only with respect to a limited number non-pathogenic indicator strains (Micrococcus flavus or Streptococcus thermophilus) (Cotter et al., 2005a; Lubelski et al., 2007; Siezen et al., 1996). A recent study by Rink et al, involving the randomization of an N-terminal domain fragment of the nisin peptide, reported enhanced IC50s against specific indicator strains (Rink et al., 2007b). These, and indeed the majority of bioengineered peptides generated to date, have resulted from site-directed approaches, with random bioengineering of the intact nisin peptide have been carried out only on a relatively small scale only and furthermore, have been largely unsuccessful, yielding derivatives with reduced or absent antimicrobial activity.
Nisin A is a cationic antimicrobial peptide due to the presence of 5 positively charged residues (Lys12 , Lys22 , Lys34 , His27 , His31) and the absence of negatively charged residues. The consequences of charge manipulation to date have been variable. Yuan et al, 2004 disclosed that the incorporation of negatively charged residues had a detrimental impact (e.g. the hinge mutants N20E, M21E and K22E) and subsequently revealed that the introduction of positively charged residues had a more beneficial outcome with respect to anti-gram-negative activity, (N20K and M21K). Introduction of a glutamic acid into the hinge region also resulted in loss of bioactivity for M21E, and a lack of any detectable production of a K22E peptide.
A further unusual feature of the nisin lantibiotic is the absence of aromatic residues. To date all aromatic residue-containing forms of nisin have been bioengineered derivatives and all have displayed reduced antimicrobial activity (i.e. I1W, M17W, V32W, 130W, N20F and N20F/M21L/K22Q (Breukink et al., 1998; Martin et al., 1996; Yuan et al., 2004). Hasper et al, and Widermann et al, both established that proline incorporation i.e. N20P/M21P, resulted in the generation of a nisin peptide incapable of pore formation. A number of small amino acids have previously been introduced into the hinge region of nisin Z, including M21G (slightly reduced activity), K22G (slightly reduced activity) and N20A/K22G (as part of an epidermin-like hinge N20A/M21K/Dhb/K22G, greatly reduced activity; (Yuan et al., 2004). It has further been reported by Yuan et al, that an N20Q substitution in nisin Z results in slightly diminished activity (Yuan et al., 2004).
The recognition of resistance to lantibiotics such as nisin is growing, a factor, which contributes to the urgent need for alternative antimicrobial peptides that exhibit a superior antimicrobial activity towards gram-positive bacteria. There is thus a need for additional anti-microbial agents, which would be effective against strains that are insensitive to nisin A or other nisin forms.
The current inventors have generated randomly mutated nisin peptides and screened for enhanced antipathogen activity. This approach, coupled to subsequent site-directed and site-specific saturation mutagenesis, yielded the identification of derivatives of nisin with enhanced activity against specific gram-positive pathogens.
Nisin is a 34 amino acid peptide and the nomenclature use for the derivatives relate to the position of the amino acid being altered. The two peptides that are already known to possess enhanced anti-gram positive activity are T2S and M17Q/G18T i.e. a single and double mutant, respectively, of nisin Z (i.e. a natural variant of the nisin A used herein which differs by having an N rather than H residue at position 27). The current derivatives are derived from nisin A and also differ with respect the location and nature of the changes i.e. positions 20, 21 and 22.
The previously generated peptides possess enhanced activity against the non-pathogenic indicator strains Micrococcus flavus and/or Streptococcus thermophilus only.
Potentially the derivatives of the invention find use in any infection involving bacteria including blood stream infections, lesions, ulcers, dental plaque, bad breath, diseases of the colon, diarrheal disease, gastric disorders associated with Helicobacter pylori, to name a few.